Comparing a Linguistic and a Stochastic Tagger

Abstract

Concerning different approaches to automatic PoS tagging: EngCG-2, a constraint-based morphological tagger, is compared in a double-blind test with a state-of-the-art statistical tagger on a common disambiguation task using a common tag set. The experiments show that for the same amount of remaining ambiguity, the error rate of the statistical tagger is one order of magnitude greater than that of the rule-based one. The two related issues of priming effects compromising the results and disagreement between human annotators are also addressed.

8 pages, LaTeX, 2 postscript figures. E-ACL'97

References in corpus (3)

• Tagging accurately -- Don't guess if you know
• A syntax-based part-of-speech analyser
• Handling Sparse Data by Successive Abstraction

Cited by in corpus (5)

• Comparing a statistical and a rule-based tagger for German
• Keyphrase Extraction using Sequential Labeling
• On the Evaluation and Comparison of Taggers: The Effect of Noise in Testing Corpora
• The DeLiVerMATH project - Text analysis in mathematics
• Development of POS tagger for English-Bengali Code-Mixed data